Matching entries: 0
settings...
AuthorTitleYearJournal/Proceedings
Nusser, Sebastian; Otte, Clemens; Hauptmann, Werner & Kruse, Rudolf Learning Verifiable Ensembles for Classification Problems with High Safety Requirements 2010 Intelligent Soft Computation and Evolving Data Mining: Integrating Advanced Technology (pp. 405-431)
IGI Global.  
BibTeX:
@incollection{Nusser2010IGI,
  author = {Sebastian Nusser and Clemens Otte and Werner Hauptmann and Rudolf Kruse},
  title = {Learning Verifiable Ensembles for Classification Problems with High Safety Requirements},
  booktitle = {Intelligent Soft Computation and Evolving Data Mining: Integrating Advanced Technology},
  editor = {Leon S. L. Wang and Tzung-Pei Hong},
  publisher = {IGI Global},
  year = {2010},
  chapter = {19},
  pages = {405-431},
  doi = {https://doi.org/10.4018/978-1-61520-757-2.ch019}
}
Nusser, Sebastian; Otte, Clemens; Hauptmann, Werner; Leirich, Oskar; Krätschmer, Manfred & Kruse, Rudolf Maschinelles Lernen von validierbaren Klassifikatoren zur autonomen Steuerung sicherheitsrelevanter Systeme 2009 at - Automatisierungstechnik Vol. 57 (3) (pp. 138-145)
Oldenbourg Verlag. (in German) 
BibTeX:
@article{Nusser2009at,
  author = {Sebastian Nusser and Clemens Otte and Werner Hauptmann and Oskar Leirich and Manfred Krätschmer and Rudolf Kruse},
  title = {Maschinelles Lernen von validierbaren Klassifikatoren zur autonomen Steuerung sicherheitsrelevanter Systeme},
  journal = {at - Automatisierungstechnik},
  publisher = {Oldenbourg Verlag},
  year = {2009},
  volume = {57},
  number = {3},
  pages = {138-145},
  note = {in German},
  doi = {https://doi.org/10.1524/auto.2009.0761}
}
Nusser, Sebastian; Otte, Clemens & Hauptmann, Werner Multi-Class Extension of Verifiable Ensemble Models for Safety-Related Applications 2009 Advances in Data Analysis, Data Handling and Business Intelligence
Studies in Classification, Data Analysis, and Knowledge Organization. (pp. 733-744)
Springer.  
BibTeX:
@inproceedings{Nusser2009GfKl,
  author = {Sebastian Nusser and Clemens Otte and Werner Hauptmann},
  title = {Multi-Class Extension of Verifiable Ensemble Models for Safety-Related Applications},
  booktitle = {Advances in Data Analysis, Data Handling and Business Intelligence},
  editor = {Andreas Fink and Berthold Lausen and Wilfried Seidel and Alfred Ultsch},
  series = {Studies in Classification, Data Analysis, and Knowledge Organization},
  publisher = {Springer},
  year = {2009},
  pages = {733-744},
  doi = {https://doi.org/10.1007/978-3-642-01044-6_67}
}
Nusser, Sebastian Robust Learning in Safety-Related Domains - Machine Learning Methods for Solving Safety-Related Application Problems 2009 School: Otto-von-Guericke-Universität Magdeburg  
Abstract: Today, machine learning methods are successfully deployed in a wide range of applications. A multitude of different learning algorithms has been developed in order to solve classification and regression problems. These common machine learning approaches are regarded with suspicion by domain experts in safety-related application fields because it is often infeasible to sufficiently interpret and validate the learned solutions. Especially for safety-related applications, it is imperative to guarantee that the learned solution is correct and fulfills all given requirements. The basic idea of the approaches proposed within this thesis is to solve high-dimensional application problems by an ensemble of simple submodels, each of which is allowed to only use two or three dimensions of the complete input space. The restriction of the dimensionality of the submodels allows the visualization of the learned models. Thus a visual interpretation and validation according to the existing domain knowledge becomes feasible. Due to the visualization, an unintended and possibly undesired extra- and interpolation behavior can be discovered and avoided by changing the model parameters or selecting other submodels. Since the learned submodels are interpretable the correctness of the learned solution can therefore be guaranteed. The ensemble of the submodels compensates for the limited dimensionality of the individual submodels. The proposed ensemble methods are successfully applied on common benchmark problems as well as on real-world application problems with very high requirements on the functional safety of the learned solution.
BibTeX:
@phdthesis{Nusser2009PhD,
  author = {Sebastian Nusser},
  title = {Robust Learning in Safety-Related Domains - Machine Learning Methods for Solving Safety-Related Application Problems},
  school = {Otto-von-Guericke-Universität Magdeburg},
  year = {2009},
  month = {July},
  url = {http://d-nb.info/999803743/34}
}
Nusser, Sebastian; Otte, Clemens & Hauptmann, Werner Verifiable Ensembles of Low-Dimensional Submodels for Multi-Class Problems with Imbalanced Misclassification Costs 2009 Applications of Supervised and Unsupervised Ensemble Methods
Studies in Computational Intelligence. Vol. 245 (pp. 191-211)
Springer.  
BibTeX:
@incollection{Nusser2009SUEMA,
  author = {Sebastian Nusser and Clemens Otte and Werner Hauptmann},
  title = {Verifiable Ensembles of Low-Dimensional Submodels for Multi-Class Problems with Imbalanced Misclassification Costs},
  booktitle = {Applications of Supervised and Unsupervised Ensemble Methods},
  editor = {Oleg Okun and Giorgio Valentini},
  series = {Studies in Computational Intelligence},
  publisher = {Springer},
  year = {2009},
  volume = {245},
  chapter = {11},
  pages = {191-211},
  doi = {https://doi.org/10.1007/978-3-642-03999-7_11}
}
Nusser, Sebastian; Otte, Clemens & Hauptmann, Werner Interpretable Ensembles of Local Models for Safety-Related Applications 2008 Proceedings of 16th European Symposium on Artificial Neural Networks (pp. 301-306)
D-facto publications.  
Abstract: This paper discusses a machine learning approach for binary classification problems which satisfies the specific requirements of safety-related applications. The approach is based on ensembles of local models. Each local model utilizes only a small subspace of the complete input space. This ensures the interpretability and verifiability of the local models, which is a crucial prerequisite for applications in safety-related domains. A feature construction method based on a multi-layer perceptron architecture is proposed to overcome limitations of the local modeling strategy, while keeping the global model interpretable.
BibTeX:
@inproceedings{Nusser2008ESANN,
  author = {Sebastian Nusser and Clemens Otte and Werner Hauptmann},
  title = {Interpretable Ensembles of Local Models for Safety-Related Applications},
  booktitle = {Proceedings of 16th European Symposium on Artificial Neural Networks},
  editor = {Michel Verleysen},
  publisher = {D-facto publications},
  year = {2008},
  pages = {301-306},
  url = {http://www.dice.ucl.ac.be/Proceedings/esann/esannpdf/es2008-34.pdf}
}
Nusser, Sebastian; Otte, Clemens & Hauptmann, Werner An EM-based Piecewise Linear Regression Algorithm 2008 Proceedings of 3rd International Workshop on Hybrid Artificial Intelligence Systems
Lecture Notes in Artificial Intelligence. Vol. 5271 (pp. 466-474)
Springer.  
Abstract: This contribution describes an EM-like piecewise linear regression algorithm that uses information about the target variable to determine a meaningful partitioning of the input space. The main goal of this approach is to incorporate information about the target variable in the prototype selection process of a piecewise regression approach. Furthermore, the proposed approach is designed to provide an interpretable solution by restricting the dimensionality of the local regression models. We will show that our approach achieves a similar predictive performance on benchmark problems compared to standard regression methods - while the model complexity of our approach is reduced.
BibTeX:
@inproceedings{Nusser2008HAIS,
  author = {Sebastian Nusser and Clemens Otte and Werner Hauptmann},
  title = {An EM-based Piecewise Linear Regression Algorithm},
  booktitle = {Proceedings of 3rd International Workshop on Hybrid Artificial Intelligence Systems},
  editor = {Emilio Corchado and Ajith Abraham and Witold Pedrycz},
  series = {Lecture Notes in Artificial Intelligence},
  publisher = {Springer},
  year = {2008},
  volume = {5271},
  pages = {466-474},
  doi = {https://doi.org/10.1007/978-3-540-87656-4_58}
}
Nusser, Sebastian; Otte, Clemens & Hauptmann, Werner Multi-Class Modeling with Ensembles of Local Models for Imbalanced Misclassification Costs 2008 Proceedings of 2nd Workshop on Supervised and Unsupervised Ensemble Methods and Their Applications (at ECAI 2008) (pp. 36-40)  
Abstract: In this paper, we will discuss different strategies of extending an ensemble approach based on local binary classifiers to solve multi-class problems. The ensembles of binary classifiers were developed with the objective of providing interpretable local models for use in safety-related application domains. The ensembles assume highly imbalanced misclassification costs between the two classes. The extension to multi-class problems is not straightforward because common multi-class extensions might induce inconsistent decisions. We propose an approach that avoids such inconsistencies by introducing a hierarchy of misclassification costs. We will show that by following such a hierarchy it becomes feasible to extend the binary ensemble and to achieve a good predictive performance.
BibTeX:
@inproceedings{Nusser2008SUEMA,
  author = {Sebastian Nusser and Clemens Otte and Werner Hauptmann},
  title = {Multi-Class Modeling with Ensembles of Local Models for Imbalanced Misclassification Costs},
  booktitle = {Proceedings of 2nd Workshop on Supervised and Unsupervised Ensemble Methods and Their Applications (at ECAI 2008)},
  editor = {Oleg Okun and Giorgio Valentini},
  year = {2008},
  pages = {36-40},
  url = {http://eprints.pascal-network.org/archive/00004311/01/SUEMA_Proceedings.pdf#page=37}
}
Nusser, Sebastian; Otte, Clemens & Hauptmann, Werner Learning Binary Classifiers for Applications in Safety-Related Domains 2007 Proceedings of 17th Workshop Computational Intelligence (pp. 139-151)
Universitätsverlag Karlsruhe.  
Abstract: This paper introduces two binary classification approaches to cope with the specific requirements of safety-related application problems. Both approaches are based on ensembles of local models, each utilizing only a small subspace of the complete input space. Thus, the interpretability and verifiability of the local models is ensured, which is a crucial prerequisite for applications in safety-related domains. Similar to the Boosting approaches, the ensemble of local models enhances the predictive performance of the final global model - compared to the limited predictive performance of the local models.
BibTeX:
@inproceedings{Nusser2007,
  author = {Sebastian Nusser and Clemens Otte and Werner Hauptmann},
  title = {Learning Binary Classifiers for Applications in Safety-Related Domains},
  booktitle = {Proceedings of 17th Workshop Computational Intelligence},
  editor = {Ralf Mikut and Markus Reischl},
  publisher = {Universitätsverlag Karlsruhe},
  year = {2007},
  pages = {139-151},
  url = {http://digbib.ubka.uni-karlsruhe.de/volltexte/1000009271}
}
Otte, Clemens; Nusser, Sebastian & Hauptmann, Werner Machine Learning Methods for Safety-Related Domains: Status and Perspectives 2006 Proceedings of Symposium on Fuzzy Systems in Computer Science (pp. 139-148)  
Abstract: Machine learning provides significant advantages concerning data-driven modelling. However, these strengths are finding only slow acceptance in safety-related applications. This paper gives a survey of machine learning in safety-related domains and discusses methods to validate the safety of the learned model. Hybrid approaches play a dominant role, Neuro-Fuzzy techniques in particular. We briefly review the pros and cons of Neuro-Fuzzy and discuss a special form of a mixture-of-experts approach to meet safety requirements. We conclude with an outline of further research perspectives.
BibTeX:
@inproceedings{Otte2006,
  author = {Clemens Otte and Sebastian Nusser and Werner Hauptmann},
  title = {Machine Learning Methods for Safety-Related Domains: Status and Perspectives},
  booktitle = {Proceedings of Symposium on Fuzzy Systems in Computer Science},
  year = {2006},
  pages = {139-148},
  url = {http://fuzzy.cs.uni-magdeburg.de/fscs2006/docs/FSCS2006_Proceedings.pdf#page=149}
}
Nusser, Sebastian Multiple Testing Error Assessment for Strongly Dependent Hypotheses using Resampling 2005 School: Otto-von-Guericke-Universität Magdeburg  
BibTeX:
@mastersthesis{Nusser2005,
  author = {Sebastian Nusser},
  title = {Multiple Testing Error Assessment for Strongly Dependent Hypotheses using Resampling},
  school = {Otto-von-Guericke-Universität Magdeburg},
  year = {2005},
  month = {September},
  url = {http://www.seb-nusser.de/paper/Diplomarbeit.pdf}
}
Nusser, Sebastian Inductive Causation based on Binary Data 2004 School: Otto-von-Guericke-Universität Magdeburg  
BibTeX:
@techreport{Nusser2004,
  author = {Sebastian Nusser},
  title = {Inductive Causation based on Binary Data},
  school = {Otto-von-Guericke-Universität Magdeburg},
  year = {2004},
  month = {October},
  type = {Bachelor's Thesis},
  url = {http://www.seb-nusser.de/paper/Studienarbeit.pdf}
}